A Hardware Acceleration System Oriented to LSTM Network Model

A network model and hardware acceleration technology, applied in biological neural network models, climate sustainability, neural architecture, etc., can solve the problems of lack of research results and general optimization effect at the computing level, so as to achieve high resource occupation and improve on-chip memory access Efficiency, the effect of reducing memory access time
CN113191488BActive Publication Date: 2022-05-20HUAZHONG UNIV OF SCI & TECH
5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN ยท China
Patent Type
Patents(China)
Current Assignee / Owner
HUAZHONG UNIV OF SCI & TECH
Publication Date
2022-05-20

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a hardware acceleration system oriented to an LSTM network model, and belongs to the technical field of deep learning hardware acceleration. The invention discloses a hardware acceleration system oriented to a deep learning long short-term memory (LSTM) network model, which includes a network reasoning calculation core and a network data storage core. As the computing accelerator of the LSTM network model, the network reasoning computing core deploys computing units according to the network model to realize the computing acceleration of convolution operations, matrix point multiplication, matrix addition, activation functions and other computing units; the network data storage core serves as the data of the LSTM network model The cache and interaction controller deploys the on-chip cache unit according to the network model to realize the data interaction link between the computing core and the off-chip memory. The invention improves the calculation parallelism of the LSTM network model, reduces the processing delay, reduces the memory access time, and improves the memory access efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the field of deep learning hardware acceleration, and more specifically relates to a hardware acceleration system oriented to an LSTM network model. Background technique

[0002] Long Short-Term Memory (LSTM), as a variant of deep learning Recurrent Neural Network (RNN), is widely used in sequence model processing tasks such as speech recognition, natural language processing, and image compression. LSTM effectively solves the problem of gradient explosion and gradient disappearance in the RNN training process by introducing a gating mechanism and a state value for storing long-term and short-term historical information, and relatively greatly increases its computational complexity and space complexity. Its intensive calculation and memory access limit its application on the embedded hardware platform with limited resources. Therefore, it is a very meaningful research to design and accelerate the optimization of the LSTM model f...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More